This paper provides a critical analysis of some of the key features of merger policy as understood and practiced in leading jurisdictions such as the European Community and the United States. It focuses first on a discussion of the gradual move of merger policy towards the examination of unilateral effects. The critical appraisal of this process is based on the practical and theoretical shortcomings of the economic models that underlie the growing prominence of unilateral effects as the key anticompetitive factor arising from a proposed merger. The paper stresses that even if unilateral effects were to lead to an increase in the conventional measures of anticompetitive performance (such as markups), it is not clear that this implies less competitive behavior for many of the most relevant industries in today's advanced economies. Finally, the paper also examines the relation between competition and welfare, and argues that even if competition does indeed diminish due to a merger, it is not a straightforward conclusion that this is not good in terms of economic welfare when we take fully into consideration the incentives to innovate and the dynamic welfare gains that arise from new products and production processes.